The Evolution of Glued Trussed Joists

Back to Library

Issue #13265 - August 2021 | Page #86
By Tuomo Poutanen

I-joists dominate the joist market due to their high material efficiency and low manufacturing cost. However, this product has disadvantages, including difficulty in HVAC installations and the possibility of bottom chord support only. Therefore, the market would pay a considerable premium for a product with crosswise openings and adaptable support. The glued trussed joist is such a product. It is even more material efficient than the I-joist and, in automatic assembly, the manufacturing cost is almost the same as an I-joist.

Glued trussed joists were developed in Germany some 60 years ago when two joint concepts were introduced, DSB and PERI. [For figures, See PDF or View in Full Issue.] The DSB joist was simple – the joint fingers were not tapering and the webs were not connected to each other. The product was on the market for only about 20 years.

The PERI joist is like the DSB joist, but there is a finger joint between the webs. The PERI is used as a joist for concrete molding. In fact, the PERI is still exploited extensively on the market.

A more improved product with a finger joint between the webs and between the chord-web joint was developed in America some 30 years ago. [For figures, See PDF or View in Full Issue.] This joist has established a solid position in the market, but the success has been modest regarding the potential of the glued trussed joist.

Glued Roof Trusses

In Finland, two glued trusses for roofs were manufactured from 1980 to 2000, the IU truss and the A-truss. The IU truss was made of double plank chords and plank webs fixed between the plank chords. [For figures, See PDF or View in Full Issue.] There was a thin plywood in the chord-web joint which improved the joint resistance. This embodiment received a patent. The buckling, especially the upper chord buckling but also the web buckling, was a serious problem in this truss, however. Both chord planks must be fixed effectively against buckling. Because some collapses occurred due to buckling, production of the IU truss ended, even though it had been very competitive in the market.

The A-truss is like a connector truss, but the connector plates are replaced with plywood glued to chords and webs. [For figures, See PDF or View in Full Issue.] Although the A-truss is more expensive than the connector truss, the product was on the market for two decades. The A-joist fabricator specialized in complicated cases which the connector truss fabricators opted not to do. One may think that the resistance of the A-truss is doubtful, but this author does not know of one failure with these trusses.

Glued Timber Truss Development in Finland

During recent decades, extensive glued trussed joist development has taken place in Finland and some half dozen joint and joist types have been developed and tested. These tests have resulted in glued trussed joist commercialization in Finland two years ago. The latest joist and joint innovations are presented here. [For photos from Poutanen’s glued timber truss tests during recent decades and additional figures, See PDF or View in Full Issue.]

New Innovations

Numerous revisions to the existing technology have been made. Billet production with arbitrary cut and trim has been developed. Automatic custom joist production has been developed. The new joist can be supported arbitrarily at the bottom chord, at the top chord (without a hanger), or between the chords. The new joist has no open routings and no water pockets. The web finger is long, more than 2”, which secures a good web-chord resistance. The joist is simple – there is one web type only. The robust joist with a 3.5” chord has two fingers and the light joist with a 2.5” chord has one finger only. The finger routing tool is a saw blade which ensures high quality gluing surfaces without open routings. The web finger punches the chord which enables the maximum web-chord glue areas. The quality assurance is based on visual inspection and proof samples instead of proof loading of every joist. The new joist has full resistance symmetry, making it possible to turn the joist upside down. The high strength is uniform along the joist and every joint has high resistance.

Material efficiency

A rule of thumb is that an I-joist has about 35% lower material cost than the solid joist, LVL, and glulam. The glued trussed joist has even lower material cost than the I-joist. The glued trussed joist’s material advantage increases with increased depth. The trussed joist has a little higher manufacturing cost than the I-joist, but due to the lower material cost the factory prices are almost the same. In depths higher than about 14”, the glued trussed joist has a lower factory price assuming that the OSB price is at least about 50% higher than the lumber price.

Billet production

The current glued trussed joists are manufactured either to custom lengths or to standard lengths with a trimming option on one joist end. An innovation is that the trussed joist is manufactured as actual billet, e.g., in 40’ length. Arbitrary cutting and trimming are possible, e.g., the billet can be cut to multiple workable joists. If the joist is cut at a node of the trussed joist, no reinforcement in the cutting point is needed. However, if the joist is cut between the nodes, a reinforcement such as OSB panels nailed on the chord edges is needed. [For image, See PDF or View in Full Issue.]

Cut and trim between the nodes

Each support option – top chord, bottom chord, and between the chords – needs different reinforcement. Figure 8 shows the basic schemes [for images, See PDF or View in Full Issue]. In a top chord support, steel U-profile + two OSB panels are nailed to the chords. In a bottom chord support, two OSB panels nailed to the chords is sufficient. If the support is between the chords, profiled OSB panels are nailed to the chords.

Automatic manufacturing of custom joists

Glued trussed joists have a standard node distance of 24”. Custom manufacturing means that the joist length is not a multiple of 24”. A custom length is obtained by removing one web in the middle of the joist, as shown in Figure 9 [for image, See PDF or View in Full Issue]. At each end of the joist, the node distance is 24”.

Routing tool

The finger routings in the current glued trussed joists are made by using a cursor. In the new joint, a normal saw blade is used. Such routing is a little slower, but the routing result is good regarding gluing surfaces and knots and other defects do not cause any harm. Furthermore, the joint has no open routings and no water pockets. The finger normally is tapering, which means that each routing must be made in two steps, one for the first tapering side and another one for the second.

Finger length, glue area

In the new joint, the web finger punches the chord, and the web-chord glue area is the maximum possible, i.e., the full overlapping area of the web flat and the chord edge. The finger length is more than 2” and more than double regarding the finger length in the current joints and the web-chord glue area is about 50% higher. [For images, See PDF or View in Full Issue.] The shear stress in the gluing surface is low, which results in high joint strength.

Defects, knots

Glued timber trussed joists are normally made of sawn timber which includes knots and other defects. A knot in a glued timber joint considerably worsens the joint strength, even to the point of the strength vanishing. It is possible to overcome this deficiency, however, by making the joint big, bigger than any feasible knot. In the new joist, the joint is big, the biggest possible, i.e., the overlapping area of the web flat and the chord edge. The joint area in the new joist is about three times larger than in the current glued trussed joists. If there is a knot in the joint, it never covers the whole joint. There is always sufficiently sound timber to obtain the required resistance.

Proof loading, quality assurance

The quality assurance of the current glued trussed joists is based on proof loading of each joist brought to the market. The quality assurance of the new joist is like the quality assurance of glulam and I-joists. The quality assurance is based on small test samples and visual inspection. Proof loading is applied occasionally only.

Full resistance symmetry and consistency

The new joist is fully symmetric – it can be turned upside down and the strength is consistent in the whole length. These characteristics are crucial in a joist manufactured as a billet and cut for the actual need because any joint in the billet may be critical, i.e., all joints must carry the characteristic design resistance of the joist.

No custom engineering needed in openings

As all joints of the joist have the same design resistance, the opening in a floor or roof can have any form and location, even round openings are possible. One standardized design covers all cases, i.e., custom design for each case is not needed. Figure 11 [for image, See PDF or View in Full Issue] shows the principle.

As glued trussed joists have evolved, it is clear that they have become more efficient and appealing in the timber industry. Their low manufacturing cost and high material efficiency make them attractive especially for floors, and they obviously will become the de facto standard joists in floors. This article addresses light floor joists, but the glued timber trusses are feasible for roofs, robust girders, and long spans up to 100 ft, too.

 

A professor of Civil Engineering for the Faculty of the Built Environment at Tampere University in Finland, Tuomo Poutanen specializes in structural design, product development in timber engineering, structural probability theory, and structural codes. After graduating from Tampere University, Poutanen founded a structural consulting office specializing in trusses. His office designed almost one half of the connector trusses in Finland during 1974–1990. A unique connector plate invented and patented by Poutanen in 1978 with barbs in the teeth is used in virtually all timber trusses manufactured in Finland. In the 1980s, Poutanen had his own truss laboratory where he developed a new connector truss design concept which considers the semirigid joint behavior and the eccentricities. Poutanen defended his dissertation about this design model in 1995, and the model is now used widely across the world. Poutanen’s research has focused in recent decades on glued timber joints and trusses. Questions about this article may be sent to tuomo.poutanen@tuni.fi.

You're reading an article from the August 2021 issue.

Search By Keyword

Issues

Book icon Read Our Current Issue

Download Current Issue PDF